nerdexam
Amazon

MLA-C01 · Question #167

A company is training a new ML model to replace a model that is deployed on an Amazon SageMaker AI real-time endpoint. An ML engineer needs to determine the latency and the accuracy of the new…

The correct answer is D. Perform shadow testing with a traffic sampling percentage of 100%. Shadow testing sends a copy of production traffic to the new model while keeping all user-facing responses served by the existing model, allowing accurate measurement of latency and accuracy in real production conditions without impacting users.

Deployment and Orchestration of ML Workflows

Question

A company is training a new ML model to replace a model that is deployed on an Amazon SageMaker AI real-time endpoint. An ML engineer needs to determine the latency and the accuracy of the new model. The ML engineer must evaluate the new model in a production scenario without affecting the users of the existing model. Which solution will meet these requirements?

Options

  • APerform a blue/green deployment with linear traffic shifting.
  • BPerform a blue/green deployment with canary traffic shifting.
  • CPerform a rolling deployment with a rolling batch size of 50% of the current fleet.
  • DPerform shadow testing with a traffic sampling percentage of 100%.

How the community answered

(36 responses)
  • A
    8% (3)
  • B
    3% (1)
  • C
    6% (2)
  • D
    83% (30)

Explanation

Shadow testing sends a copy of production traffic to the new model while keeping all user-facing responses served by the existing model, allowing accurate measurement of latency and accuracy in real production conditions without impacting users.

Topics

#Shadow testing#Model deployment strategies#Real-time inference#Model evaluation

Community Discussion

No community discussion yet for this question.

Full MLA-C01 Practice